Das Experiment & seine Ziele (Hintergrundinformation für PR) Bernhard Schmidt DESY März 2002

- Fixed target Experiment, benutzt nur die HERA Protonen - Vorwärtsspektrometer mit Teilchenidentifizierung 12 -200 mrad ~ 85% von 4p - hohe Wechselwirkungsraten (107 pro Sekunde)

Internes Drahttarget im Vakuumsystem des HERA-p Rings stable routine operation , IR 5 - 40 MHz basically smooth cohabitation with ep experiments Carbon Aluminium Titanium Palladium Tungsten Target wires inside vacuum vessel

~ 5•1013 inelastic interactions ! Produced Luminosity ~ 5•1013 inelastic interactions ! Rate fluctuations 10 % @ 1mm Sun Mo Fr Sa

Das Spektrometer : Dipolmagnet + Spurdetektoren Bdl=2 Tm Vertex Tracker Main Tracker

The Vertex Tracking System 64 double sided Si µ-strip detectors in 8 super layers 150 000 read out channels Roman pot system in vacuum tank Detector modules movable radial and laterally by manipulators

Detector modules Steel bands Caps removed

The VDS works routinely and close to design specs  = 50 µm z = 500 µm impact parameter resolution P [GeV] Distribution of reconstructed Primary vertices on 8 wires

Main Tracker (gas filled detectors) 3•107 particles / sec • 1/r2 tracking in high flux environment ITR OTR rmin = 6 cm < 106 particles cm-2 sec-1 forward hemisphere in CM rmin = 20 cm < 105 particles cm-2 sec-1 backward hemisphere in CM

I T R Inner Tracker The world largest (gas) micro pattern tracker 184 individual chambers 27 x 27 cm2 each - 18 m2 detector surface 140 000 read out channels (ADC) Construction of ITR chambers - two step gas amplification Micro Strip Gas Chambers Gas Electron Multiplier 300 µm strip pitch on glass substrate only 6 mm total height read out by custom made ASIC chips (HELIX) - neue Technologie ! CMS, LHC-b -> no

First ‘real’ experience in 2002 Run time experience in 2000 - no routine operation so far - need careful training - large gain variations between chambers - individual gain adjustment needed - no trigger signals due to large feedback noise Performance figures efficiency >90% seems possible (design 98 %) resolution ~80µm (at design) HV stability fine, no problems at high rates First ‘real’ experience in 2002

O T R OTR The world largest honeycomb tracker 1000 individual modules max. of superlayer 4.6 x 6.5 meters - 1000 m2 detector surface 115 000 read out channels (TDC) O T R Thin, lightweight (X0) construction self sustaining mass producible affordable Construction of OTR modules open honeycomb geometry 5 mm and 10 mm drift cells produced layer by layer wires supported by FR4 bridges no forced gas flow

OTR mass production was a big enterprise .. ..our courageous young colleagues faced the challenge .. 1.5 Mio solder points on wires ….! 1ooo modules produced in 9 months ! > 100 physicists and technicians working in parallel on 4 different places Peking Dubna Zeuthen Hamburg OTR World 4 6 5 8 Module Mass Production work places

All OTR superlayers installed by end 1999.

OTR routinely used for tracking in 2000 Performance and problems : HV stability at the limit voltage ~4% reduced compared to test beam substantially improved during shutdown, 16 000 capacitors replaced Additional noise from TDC - trigger connection Thresholds 4 fC instead of 2.5 fC big improvement during shutdown, new drivers, optimisation of cabling etc.

Teilchenidentifizierung RICH ECAL µ-Detektor p 20 m

RICH detector Particle Separation NO Cherenkov relation for q2[mrad] 2 Read out by multi-anode phototubes NO TAMEA chambers CsI cathodes for high rate environment !  Particle Separation K p p 1/p2[GeV/c]-2 Very stable in 2000 # photons, resolution at design

ECAL electron pre-triggers photon detection

p0 2000 largely completed some problems with noise, stability, hot channels.. …. used for online calibration Shutdown : Modified analogue read out with better S/N h Draw back : HERA-B got fatter than foreseen dead material! > 1 X0 in front of ECAL Shutdown : All tracking chambers in Magnet removed

Muon detector µ pre-trigger drift tubes + iron absorber 4 superlayers 3 types of chambers ~15 000 cha. Tube chambers (342) : 2 cm rectangular aluminium profile as drift tubes Pad chambers (132): as tube chambers, additional cathode pads for fast trigger signals ~8 000 cha. ~5 300 cha. Pixel chambers (16): gas pixel chambers,short wires parallel to beam. Covering the innermost part

Shutdown work : Experience in 2000 run : Muon identification tube chambers work fine with efficiency close to design gas changed to CH4-free mixture due to anticipated aging problems pixel chambers ok, initial noise problems solved but not used for tracking so far (missing ITR) main worry : problems with system noise, especially for pad read out ! Muon identification + tracking : ->mm2.5 4 ->mm  ->mm1.8 4 low pre-trigger efficiency ! Shutdown work : chambers taken out and minor problems fixed (gas leaks, electronics problems) pad system completely dismounted and overhauled noise situation improved by better grounding scheme

The HERA-B DAQ and Trigger System CERN Courier 200 128 ev. deep pipeline Suppression factor 1.7 Gbyte/sec 100 240 events processed simultaneously LINUX farm, highly flexible 300/450 MHz->1.4 GHz Full event reconstruction in 240 nodes LINUX farm 10 all hardware set up all but L1-trigger used at design performance in 2000 suppression factor 2•105

Lepton pair trigger 2000 : stuck in commissioning phase First Level Trigger & pre-trigger Lepton pair trigger starting from ‘pre-trigger seeds’ electrons muons following lepton track candidates through chambers -> trigger tracks combining track pairs applying mass cut ( > 2 GeV) in less than 12 µsec network of custom made processors interconnected by high speed optical links 2000 : stuck in commissioning phase

Das Spektrometer funktioniert …. (2000er Daten) rekonstruierte (seltsame) Zerfälle Ks0 L pT [GeV/c] pL - asymmetry

Di-lepton trigger Elektronen Muonen

CP CP violation in B - decays B0 J/yKs p+p- e+e- m+m- Wozu der Aufwand ??? CP The initial goal CP violation in B - decays The ‘golden decay’ is rare ! BR = 5•10-4 1 out of ~25 000 B0 BR = 0.7 B0 J/yKs p+p- e+e- m+m- CP measurement needs > 1000 golden decays BR = 0.12

& HERA is not a very good place to produce B’s… sB/tot=10-6 rareness of decay : 4•10-5 impurity of B sample : 10-6 Signal/BG = 4•10-11 ! To see CP : ~1000 clean events needed with ~ 25 % efficiency observe ~1014 events ! 107 seconds • 107 events / second detector standing high rates for long time sophisticated trigger event selection data acquisition &

das hat nicht so ganz funktioniert… HERA-B hat an vielen Stellen technologisches Neuland betreten Die dabei aufgetretenen Probleme waren DEUTLICH mehr und gravierender als von den Experten erwartet Das Zeitplan für HERA-B hatte keinerlei Reserven Das Konzept von HERA-B ist extrem anfällig, auch gegen kleine Unzulänglichkeiten , Erfolg = (Einzelperfektion)n…… ABER : HERA-B hat enorme technologische Einsichten und Erkenntnisse gebracht …-> LHC-Experimente

Wie geht's weiter ? Was macht man jetzt mit dem Detektor ? hochspezialisiertes Experiment : lepton pair trigger, mll > 2 GeV (im wesentlichen J/) Viel häufiger als J/aus B-Zerfällen sind direkte J/ Was ist daran interessant ?? Charmonium Produktion in Kernen ! p+N -> J/X p+N -> ’X p+N -> ’X s cc =s0•Na Als Funktion der Massenzahl N des Targets .. a ≠ 1 “suppression”

Positive xF : nuclear medium sees only pre.-res. state ratio Y’/(J/Y) should be A-independent (universal suppression coefficient) Negative xF : nuclear medium sees fully formed charmonium states suppression very different for J/Y‘, cc ! a Here data exist Here we are ! Grosse Winkel ! Here NO data exist xF

HERA-B 2002 / 2003 Firsts result from 2001 commissioning run measurement of  B -> J/yX J/y  ~ 30 % (mainly systematic) 2002 : resume regular data taking - complete the commissioning understand capabilities of detector and trigger get ‘Physics’ data ! Firsts result from 2001 commissioning run Measurement of charmonium production for different target materials (C, Al, Ti, Pd, W) ‘charmonium suppression’ different targets simultaneously (relative measurements) measure J/y~1.5 M), y ‘~26 k), c ~100 k), sensitive in negative XF range (-0.3 to +0.2) observe full final state s = s0 Aa expected error on a : < 1% for each XF bin